The Kestrel MCP server exposes workflow management as tools that AI coding agents (Cursor, Claude Desktop, OpenAI Codex, etc.) can call directly. This lets you create, modify, and monitor workflows through conversational AI.
Prerequisites
Install the Kestrel CLI and authenticate:
brew install KestrelAI/tap/kestrel
kestrel auth kestrel_sk_...
~/.kestrel/config.json.
Setup
The MCP server runs as a subcommand of the CLI. Configure your AI tool to run kestrel mcp.
Cursor
Add to ~/.cursor/mcp.json:
{
"mcpServers": {
"kestrel": {
"command": "kestrel",
"args": ["mcp"]
}
}
}
Claude Desktop
Add to ~/Library/Application Support/Claude/claude_desktop_config.json:
{
"mcpServers": {
"kestrel": {
"command": "kestrel",
"args": ["mcp"]
}
}
}
OpenAI Codex
Add via the Codex CLI:
codex mcp add kestrel -- kestrel mcp
~/.codex/config.toml:
[mcp_servers.kestrel]
command = "kestrel"
args = ["mcp"]
If kestrel isn’t on your PATH, use the full path (e.g., /usr/local/bin/kestrel).
Workflow Management
| Tool | Description |
|---|
list_workflows | List all workflows, optionally filtered by status |
get_workflow | Get workflow details including definition and trigger config |
create_workflow | Create a new workflow from a definition |
update_workflow | Update a workflow’s name, description, or definition |
delete_workflow | Delete a workflow |
activate_workflow | Activate a workflow |
pause_workflow | Pause an active workflow |
duplicate_workflow | Copy a workflow as a new draft |
Workflow Generation
| Tool | Description |
|---|
generate_workflow | Generate a workflow from natural language |
request_workflow | Submit a request to trigger a matching workflow. If nothing matches, returns pending_confirmation with a request_id |
confirm_workflow_request | Resolve a pending_confirmation request: confirm to send it to the platform team, or dismiss it. The agent asks you before calling this |
list_suggested_workflows | Get AI-suggested workflows |
Execution Management
| Tool | Description |
|---|
list_executions | List executions for a workflow |
get_execution | Get execution details including step results |
cancel_execution | Cancel a running execution |
get_workflow_stats | Aggregate execution statistics |
Approval Management
| Tool | Description |
|---|
list_pending_approvals | List pending approval gates |
approve_step | Approve a gate (with optional justification) |
reject_step | Reject a gate |
Request Management
| Tool | Description |
|---|
list_unmatched_requests | List unmatched workflow requests |
approve_workflow_request | Approve a request |
reject_workflow_request | Reject a request |
Catalog and Integrations
| Tool | Description |
|---|
get_catalog | Available signal triggers and action blocks |
get_integrations_status | Integration connection status |
Integration Management
| Tool | Description |
|---|
list_integrations | List every integration with its type, required credential fields, and connection status |
connect_integration | Connect a token or knowledge-source integration, or get the browser URL for OAuth integrations (GitHub, GitLab, Slack) |
test_integration | Test a connected integration’s credentials |
disconnect_integration | Disconnect a token integration or knowledge source |
Kubernetes, AWS, and OCI use multi-step flows and must be connected with the CLI (kestrel integrations connect <name>) instead of MCP tools. For credential safety, prefer the CLI’s interactive hidden prompts for secrets rather than pasting tokens into an agent chat. Example Conversations
Creating a workflow:
“Create a workflow that triggers when a PagerDuty alert fires, runs K8s RCA, and posts results to #incidents on Slack”
The AI agent calls generate_workflow, reviews the result, then create_workflow and activate_workflow.
Checking executions:
“Show me the last 5 failed workflow executions and what went wrong”
The agent calls list_workflows, list_executions, then get_execution for failed ones.
Approving a gate:
“Are there any pending approvals? Approve the production deployment one”
The agent calls list_pending_approvals, identifies the right one, and calls approve_step.
Connecting an integration:
“Connect our PagerDuty account to Kestrel”
The agent calls list_integrations to find the required credential fields, asks you for the values, then calls connect_integration and verifies with test_integration.
Requesting a change with no matching workflow:
“Request a Redis cluster for the cache layer”
The agent calls request_workflow. When no workflow matches, it gets back pending_confirmation and asks whether you want to send the request to your platform team; based on your answer, it calls confirm_workflow_request to file or dismiss it. 
